MEGANet-W: A Wavelet-Driven Edge-Guided Attention Framework for Weak Boundary Polyp Detection

TL;DR

MEGANet-W is a novel wavelet-driven attention framework that enhances weak boundary polyp detection by integrating multi-orientation Haar wavelet edge maps into the segmentation process, improving accuracy without extra learnable parameters.

Contribution

The paper introduces a wavelet-driven edge-guided attention network with multi-orientation Haar wavelet modules and edge fusion, advancing boundary detection in medical image segmentation.

Findings

Outperforms existing methods on five datasets with up to 2.3% mIoU improvement.

Achieves higher mDice scores by 1.2%, indicating better segmentation accuracy.

No additional learnable parameters are introduced, maintaining model efficiency.

Abstract

Colorectal polyp segmentation is critical for early detection of colorectal cancer, yet weak and low contrast boundaries significantly limit automated accuracy. Existing deep models either blur fine edge details or rely on handcrafted filters that perform poorly under variable imaging conditions. We propose MEGANet-W, a Wavelet Driven Edge Guided Attention Network that injects directional, parameter free Haar wavelet edge maps into each decoder stage to recalibrate semantic features. The key novelties of MEGANet-W include a two-level Haar wavelet head for multi-orientation edge extraction; and Wavelet Edge Guided Attention (W-EGA) modules that fuse wavelet cues with boundary and input branches. On five public polyp datasets, MEGANet-W consistently outperforms existing methods, improving mIoU by up to 2.3% and mDice by 1.2%, while introducing no additional learnable parameters. This approach improves reliability in difficult cases and offers a robust solution for medical image segmentation tasks requiring precise boundary detection.